|Publication number||US7277386 B1|
|Application number||US 10/293,785|
|Publication date||Oct 2, 2007|
|Filing date||Nov 12, 2002|
|Priority date||Nov 12, 2002|
|Publication number||10293785, 293785, US 7277386 B1, US 7277386B1, US-B1-7277386, US7277386 B1, US7277386B1|
|Inventors||Dennis C Ferguson, Nischal Sheth, Ken Kuwabara|
|Original Assignee||Juniper Networks, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (10), Non-Patent Citations (2), Referenced by (24), Classifications (20), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The invention relates to computer networks and, more particularly, distribution of traffic within computer networks.
A computer network is a collection of interconnected computing devices that can exchange data and share resources. In a packet-based network, the computing devices communicate data by dividing the data into small blocks called packets, which are individually routed across the network from a source device to a destination device. The destination device extracts the data from the packets and assembles the data into its original form. Dividing the data into packets enables the source device to resend only those individual packets that may be lost during transmission.
In some environments, two network devices may be connected by multiple physical interconnects. For example, two network routers may be connected by multiple Ethernet links. Similarly, multiple physical interconnects may couple other devices, e.g., two switches, a switch and a server, two routers, a switch and an end user station, and the like.
The term “link aggregation” refers to a technique by which the multiple physical interconnects are logically associated and treated as a single, aggregated link. In particular, the IEEE 802.3ad Working Group recently developed a Link Aggregation specification. This approach may provide for increased link availability and bandwidth between the two devices. For example, bandwidth can be increased in linear increments, e.g., by adding individual interconnects to increase the bandwidth between the two devices. In addition, the multiple physical interconnects provide a degree of redundancy in that a failure in a single physical interconnect no longer inhibits direct traffic flow between the two devices. With link aggregation, traffic on a failed physical interconnect can be redirected to one of the other physical interconnects in the aggregated link.
In general, link aggregation operates by inserting a layer within the networking stack between the second and third layers. For Ethernet, for example, a new Link Aggregation Layer functions between the Ethernet Media Access Control (MAC) layer, and the higher layer protocols. Each of the underlying Ethernet ports in an aggregated link transmits and receives frames with its own unique MAC address. As frames pass through the Link Aggregation layer, addresses are manipulated so the aggregated ports appear as a single link with one MAC address. This allows link aggregation to be completely transparent to higher layer protocols.
In general, the invention is directed to techniques for distributing label switched packets, such as multiprotocol label switched (MPLS) packets, associated with a common label switched path (LSP). The techniques may, for example, be used to load balance label switched packets of an LSP across an aggregated link having two or more logically associated physical interconnects. In addition, the techniques may be used to distribute label switched packets of a common LSP across multiple data paths within a network device, such as a router. As used herein, the term path refers to any physical path for communicating data. In this manner, the term encompasses both external physical interconnects between network devices, and internal data paths within an individual network device.
In one embodiment, a method comprises receiving packets associated with a common label switched path (LSP), and distributing the packets across multiple paths.
In another embodiment, a method comprises receiving a label switched packet having a label, and extracting flow information from the packet based on the label. The method further comprises distributing the packet to one of a plurality of paths based on the extracted flow information.
In another embodiment, a method comprises receiving a packet having a plurality of labels in accordance with a label switching protocol, and applying a hash function to the plurality of labels to generate a hash value. The method further comprises distributing the packet to one of a plurality of paths based on the hash value.
In another embodiment, a method comprises receiving a data unit from a network device, determining a type of communication protocol for the data unit, and selecting a label based on the type of communication protocol. The method further comprises pre-pending the selected label to the data unit to form a label switched packet in accordance with a label switching protocol, and forwarding the packet along a label switched path in accordance with the label.
In another embodiment, a network device comprises an interface card to receive packets associated with a common label switched path (LSP), and a control unit to distribute the packets across multiple paths.
In another embodiment, a computer-readable medium comprises instructions that cause a processor to receive packets associated with a common label switched path (LSP), and distribute the packets across multiple paths.
In another embodiment a computer-readable medium comprises label data that associate ranges of labels with types of payloads to control distribution of label switched packets across multiple paths.
The invention may provide one or more advantages. For example, the described techniques may allow a network device, e.g., a router, to actively distribute packets associated with a common LSP across parallel physical data paths.
As one example, the network device may apply the techniques to distribute label switched packets, such as MPLS packets, across the logically associated physical interconnects of an aggregated network link. As another example, the techniques may be used to distribute MPLS traffic across internal data paths within a network device, e.g., across multiple data paths between an input interface card and an output interface card, across internal high-speed switching components, or across multiple forwarding engines within the network device.
By distributing the MPLS packets across multiple data paths load balancing may be achieved, and the techniques may decrease packet latency through the network and the various communication devices. In addition, the techniques allow for the distribution of the MPLS traffic, while maintaining in-sequence delivery of packets of common data flows. In other words, those MPLS packets having common flow characteristics, such as MPLS packets having common source and destination addresses and ports, as well as the same communication protocol, may be distributed to the same physical interconnect in accordance with the techniques.
The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.
PE router 10A receives the MPLS packets, and forwards the MPLS packets to one or more intermediate routers (not shown) within provider network 6 for communication to PE router 10A via one of LSPs 14. PE router 10A may, for example, pre-pend additional MPLS labels to the traffic to form MPLS packets having multiple MPLS labels. In this manner, each MPLS packet traversing provider network 6 via LSPs 14 may be guaranteed to carry two or more MPLS labels.
PE router 10B receives the MPLS packets from provider network 6, and forwards the MPLS packets to CE router 8B via aggregated link 12B. PE router 10B may, for example, remove one or more MPLS labels from each of the inbound packets, and distribute the packets across the physical interconnects that are logically associated to form aggregated link 12B. Network traffic flows from customer network 4B to customer network 4A in similar manner.
In general, the devices of system 2, i.e., CE routers 8, PE routers 10, and intermediate routers (not shown) within provider network 6, utilize MPLS labels within pre-allocated ranges to convey protocol information for the payloads of the MPLS packets. In other words, the devices assign MPLS labels of the pre-allocated ranges to the MPLS packets based on the type of payloads carried by the MPLS packets. The devices may make use of this protocol information to facilitate distribution of the MPLS packets across multiple paths, e.g., multiple physical interconnects of an aggregated link.
For purposes of example, the techniques are described in reference to the MPLS protocol and MPLS packets. However, the techniques may readily be applied to any type of label switching protocol in which each packet carries a switching “label” that is associated with a defined network path, referred to herein as a “label switched path.” Accordingly, the techniques are not limited to MPLS, and may me applied to other protocols that make use of switching labels, including but not limited to ATM, Frame Relay, and the like. The term label switched path, as used herein, generally refers to any defined path of switching nodes within a network, and includes MPLS LSPs, ATM virtual circuits, and the like.
Generally, the label typically carries an index into a forwarding table, e.g., a forwarding table maintained by each of CE routers 8 and PE routers 10, which specifies the next hop for the packet. In this manner, these devices do not need to examine the label switched packets and perform next-hop routing tasks. In other words, the label associates the packet with an LSP, and carries the information that determines which path a packet should take. Forwarding decisions can be made based on a field in the packet or cell, i.e., the label, and the incoming interface, without requiring a route lookup.
One difficulty in distributing or otherwise load balancing MPLS packets across multiple physical data paths, e.g., the multiple interconnects of an aggregated link, is that an MPLS label is typically an arbitrary label unrelated to the payload. As a result, conventional mechanisms for distributing network traffic cannot identify packets belonging to a common data flow and, as a result, may distribute packets of a common data flow across separate paths. Consequently, the MPLS packets may arrive at the destination out of order, which can be problematic for the MPLS protocol, which requires in-sequence delivery. Further, the distribution mechanism may not be able to achieve substantial distribution of the traffic among the paths, which may result in congestion on one data path and unused bandwidth on another. This may decrease the efficiency of the network and increase packet latency.
In accordance with the principles of the invention, network devices of system 2 utilize the protocol information conveyed by the pre-allocated ranges of MPLS labels to control the distribution mechanism, thereby leading to improved distribution while ensuring in-sequence delivery of the packets. The devices may, for example, map the ranges of MPLS labels to respective communication protocols associated with the payloads of the MPLS packets, as illustrated in the following table:
MPLS LABEL RANGE
MULTIPLE MPLS LABELS
As illustrated in Table 1, the devices may allocate range 100000-110000, which represents an exemplary range, for MPLS packets carrying tunneled Ethernet payloads. Similarly, the devices allocate ranges of MPLS packets for carrying payloads of other types, such as Internet Protocol version 4 (IPv4), IPv6, and Asynchronous Transfer Mode (ATM) payloads. Based on the range of the MPLS label, the devices can extract a “key” from the payload that identifies the data flow of the payload carried by the MPLS packet. Using this key, the devices can readily distribute the MPLS packets across multiple paths, e.g., multiple physical links of an aggregated link.
The key extracted from the payload may include flow information, such as the source and destination IP address. The “granularity” of distribution of the MPLS packets across the paths, i.e., the extent to which MPLS packets of a common flow are distributed to a common interconnect, directly correlates with the amount of flow information included in the key. For example, application of a hashing function to a key that includes a source and destination address will result in a less granular distribution than application to a key that includes additional flow information, such as the protocol type, the source and destination port, the type of service (ToS), as well as any other information that remains constant for any one flow of data from source to destination. For payloads conforming to protocols such as ATM, the key may include information such as the ATM Virtual Channel Identifier (VCI), Virtual Path Identifier (VPI). In addition, the key may include the MPLS labels carrying the payloads, as well as other fields that may be encapsulated in an MPLS header.
In addition to defining ranges for various types of protocols, the devices may define a range of labels to indicate that the MPLS packet carries two or more MPLS labels. In Table 1 illustrated above, for example, the range 140001-150000 has been allocated for MPLS packets having multiple MPLS labels. As described in detail below, the devices may utilize the labels for distribution of the MPLS packets across multiple paths instead of extracting key data from the payloads. Because two or more MPLS labels can be used for the distribution, e.g., as inputs to a hash function, a reasonable balance of distribution may be achieved without relying on key data from the payload. This approach may be used by provider routers (not shown) within provider network 10 for internal traffic having multiple MPLS labels. This technique may offer the advantage of reduction in computing resources consumed by the distribution task, as the devices need not extract the key data from each packet.
In this manner, the devices of system 2 make use of the allocated ranges of MPLS labels to control the distribution of MPLS packets across multiple physical paths. Although the techniques have been described in reference to distribution of MPLS packets across parallel physical interconnects of an aggregated link, the techniques may be applied in a variety of environments. For example, the techniques may be readily applied to internal parallel data paths within a device. For example, the techniques may be used by a router to distribute packets across internal data paths that begin at the same input interface of the router and end at the same output interface of the router, but traverse through different internal, high-speed switch fabrics. A control unit for the router, for example, may implement the techniques to distribute received MPLS packets across multiple internal forwarding engines.
Initially, the network device receives an MPLS packet (32). For example, PE router 10A may receive an MPLS packet from provider network 6 destined for CE router 8A. The device examines the first MPLS label of the packet (34). The device determines whether the label falls within one of the pre-allocated ranges of labels (36). If the value for the label falls outside the defined ranges, the device forwards the MPLS packet across one of the available multiple paths (38) without regard to load balancing of the paths.
If the label falls within one of the defined ranges, the device performs additional processing based on the range within which the label falls. For example, if the range is mapped to a specific type of payload (40), then the device may apply a hash function using a key extracted from the payload of the MPLS packet based on the particular protocol type of the payload (42). The device then forwards the MPLS packet along one of the paths, e.g., one of the physical interconnects of an aggregated link based on the hash value (44).
If the label indicates that distribution should be applied based on multiple MPLS labels carried by the packet, the device may perform additional processing. For example, the device may apply a hash function to an MPLS label stack within the MPLS header of the packet (46), and forwards the MPLS packet based on the hash value (44).
Next, the device extracts flow information from the payload based on the protocol type, such as a payload conforming to IPv4, IPv6, Tunneled Ethernet, ATM, and other protocols (56). The device may develop a key based on the flow information. As described above, the key may contain source and destination addresses, source and destination ports, type of service (ToS), and other information specific to the data flow of the payload. The device stores the extracted key in the hash buffer (58).
After extracting the key, the device may generate a hash value by applying a hash function to the contents of the hash buffer, i.e., the key information as well as the stored MPLS label (60). After generating the hash value, the device selects one of the parallel paths, e.g., different physical interconnects of an aggregated link, based on the hash value (62). Finally, the device forwards the MPLS packet along the selected path (64). In this manner, the techniques may generate hash values having substantial distribution for different data flows associated with a common LSP, and for data flows associated with different LSPs. In addition, because MPLS packets of a common data flow have identical MPLS labels and keys, the hashing function will generate identical hash values for the MPLS packets, thereby maintaining the sequence of the MPLS packets.
Initially, the device reads the first MPLS label from the MPLS packet (66), and stores the label within an internal hash buffer (68). Next, the device determines whether all of the MPLS labels have been retrieved from the header of the MPLS packet, i.e., whether a bottom of a “stack” of MPLS labels has been reached (70).
In particular, each MPLS label includes a bottom-of-stack (BOS) bit that indicates whether the particular label is the last label within the header. If the label is not the last label, the device extracts the next label (66), and stores the label in the internal hash buffer (70). The device repeats this process until all of the labels have been read from the MPLS packet and stored in the buffer, i.e., until the BOS bit indicates that the bottom of the label stack has been reached.
Next, the device may generate a hash value by applying a hash function to the contents of the hash buffer, i.e., the plurality of stored labels (72). Because multiple labels are used as inputs to the hash function, the device may generate hash values of substantial distribution using only the MPLS labels, i.e., with out regard for to the payload of the packet. Furthermore, because MPLS packets of a common data flow will have identical labels, the techniques maintain the sequence of MPLS packets belonging to a common data flow.
After generating the hash value, the device selects one of the parallel paths, e.g., different physical interconnects of an aggregated link, based on the hash value (74). Finally, the device forwards the MPLS packet along the selected path (76).
As described, a device may selectively apply a hashing function to MPLS packet 80 based on a pre-defined range of labels within which MPLS label 82A falls. For example, the device may apply a hashing function that utilizes as inputs multiple MPLS labels 82. As an alternative, a device may use one or more fields directly from one or more of MPLS labels 82 instead of using a hashing function. As yet another example, the device may apply a hashing function that utilizes as inputs one or more of MPLS labels 82, and a key of flow information extracted from payload 84.
Next, the device accesses the packet header to determine the communication protocol for the packet, e.g., IPv4, IPv6, ATM, or the like (102). Based on the type of protocol, the device selects an appropriate MPLS label within one of the defined ranges (103), and pre-pends the label to the packet to form an MPLS packet (104).
In addition, the device extracts flow information, i.e., a key, from the payload based on the protocol type (106), and stores the extracted key in the hash buffer (108). The device generates a hash value by applying a hash function to the contents of the hash buffer, i.e., the flow information (110). After generating the hash value, the device selects one of the parallel paths, e.g., different physical interconnects of an aggregated link, based on the hash value (112).
Finally, the device forwards the MPLS packet along the selected path (114). In this manner, a router, such as CE router 8A, may generate MPLS packets at an egress for an LSP that carry labels in accordance with pre-allocated ranges. Other devices, such as PE routers 10, may make use of the labels to distribute the MPLS packets across parallel paths.
Router 120 includes a control unit 128 that directs inbound packets received from inbound links 124 to appropriate outbound links 126. In particular, the functionality of control unit 128 can be divided between a routing engine 136 and a packet-forwarding engine 130.
Routing engine 136 is primarily responsible for maintaining routing information 138 to reflect the current network topology. In order to maintain an accurate representation of the network, router 120 supports a number of protocols for exchanging routing information with other routers. For example, router 120 may support the Border Gateway Protocol (BGP), for exchanging routing information with routers of other autonomous systems. Similarly, router 120 may support the Intermediate System to Intermediate System protocol (IS-IS), which is an interior gateway routing protocol for communicating link-state information within an autonomous system. Other examples of interior routing protocols include the Open Shortest Path First (OSPF), and the Routing Information Protocol (RIP).
Routing engine 136 directs packet-forwarding engine 130 to maintain forwarding information 132 in accordance with routing information 138. Forwarding information 132 may, therefore, be thought of as derived from the information contained within routing information 138. In particular, forwarding information 132 associates packet information, e.g., a “key” as described above, with specific forwarding next hops (FNH). A FNH generally refers to a neighboring router physically coupled to router 120 along a given route. For example, the FNH for a route may specify a physical interface and media access control (MAC) address for the interface associated with the neighboring router.
Packet-forwarding engine 130 controls the flow of packets through router 120, and distributes MPLS traffic across parallel paths, as described herein. In particular, router 120 includes a plurality of interface cards (IFCs) 122 for sending and receiving packets via inbound links 124 and outbound links 126. IFCs 122 may include a number of interface ports (not shown) for coupling to network links 124 and 126.
Specifically, upon receiving an inbound packet via an input port of one of IFCs 122, packet-forwarding engine 130 selects an appropriate FNH for the packet based on forwarding information 132. Packet forwarding engine 130 maps the selected FNH to an output interface for IFCs 122, and relays the packet to the selected output interface for transmission.
For MPLS packets, packet-forwarding engine 130 distributes the MPLS packets in accordance with the techniques described herein. For example, for MPLS packets destined for an aggregated link, packet-forwarding engine 130 distributes the packets across the output interfaces associated with the multiple physical interconnects of the aggregated link. In particular, packet-forwarding engine 130 makes use of label data 140 to control distribution of the MPLS packets. For example, label data 140 may map a label of the inbound MPLS packet to a protocol type, or may designate the MPLS packet as having multiple labels, as illustrated in Table 1 above.
Packet-forwarding engine 130 includes hash buffer 124 to store an extracted key and one or more MPLS labels of the packet. Packet-forwarding engine 130 applies a hash function to the contents of hash buffer 134 to generate a hash value. Based on the generated hash value, packet-forwarding engine 130 relays the MPLS packet to one of the physical interfaces of IFCs 22 for transmission.
Packet-forwarding engine 130 may arrange label data 140 in the form of a variety of data structures, such as a number of tables, link lists, radix trees, databases, flat files, or other data structures. Furthermore, packet-forwarding engine 130 may store the data structures on one or more computer-readable media, such as a magnetic medium, optical medium, non-volatile random access memory (NVRAM), dynamic random access memory (DRAM), FLASH memory, or the like.
Each of packet-forwarding engine 130 and routing engine 136 may comprise one or more dedicated processors, discrete hardware circuitry, and the like, and may operate according to executable instructions fetched from a computer-readable medium. Examples of such media include random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, and the like. Furthermore, although the functionality has been described in reference to a central control unit of a network device, e.g., control unit 128 of router 120, the functionality may be distributed within the network device, e.g., within IFCs 122.
A number of embodiments of the present invention have been described. Furthermore, the techniques may be applied in a variety of environments. For example, the techniques may be used to distribute label switched packets across physical interconnects of an aggregated link, as described. In addition, the techniques may be readily applied to internal parallel data paths within a device, such as data paths that begin at the same input interface of a router and end at the same output interface of the router, but traverse through different interconnects. For example, a control unit for a router may implement the techniques to distribute received packets across multiple internal forwarding engines. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the invention. Accordingly, other embodiments are within the scope of the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US6091725||Dec 29, 1995||Jul 18, 2000||Cisco Systems, Inc.||Method for traffic management, traffic prioritization, access control, and packet forwarding in a datagram computer network|
|US6118760||Jun 30, 1997||Sep 12, 2000||Sun Microsystems, Inc.||Management of entries in a network element forwarding memory|
|US6243667||May 28, 1996||Jun 5, 2001||Cisco Systems, Inc.||Network flow switching and flow data export|
|US6553005 *||Jul 26, 2000||Apr 22, 2003||Pluris, Inc.||Method and apparatus for load apportionment among physical interfaces in data routers|
|US6651099||Jun 30, 2000||Nov 18, 2003||Hi/Fn, Inc.||Method and apparatus for monitoring traffic in a network|
|US6909713||Sep 5, 2001||Jun 21, 2005||Intel Corporation||Hash-based data frame distribution for web switches|
|US6980550||Jan 16, 2001||Dec 27, 2005||Extreme Networks, Inc||Method and apparatus for server load balancing|
|US20010043585||Mar 5, 2001||Nov 22, 2001||Heinrich Hummel||Method for distributing the data-traffic load on a communication network and a communication network for implementing this method|
|US20040013113||Jul 17, 2002||Jan 22, 2004||Ranjeeta Singh||Technique to improve network routing using best-match and exact-match techniques|
|US20040258073 *||Aug 14, 2001||Dec 23, 2004||Cedell Alexander||Load-sharing technique for distributing multi-protocol label switching protocol encapsulated flows across multiple physical links|
|1||U.S. Appl. No. 10/208,250, entitled "Network Traffic Distribution Across Parallel Paths," filed Jul. 29, 2002, Ross W. Callon.|
|2||*||Viswanathan et al. "Evolution of Multiprotocol Label Switching". IEEE. May 1998. pp. 165-173.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7443832 *||Dec 15, 2003||Oct 28, 2008||Alcatel||Device for determining switching paths in a label switched communication network in the presence of selection attributes|
|US7660290 *||Apr 14, 2005||Feb 9, 2010||Cloud Scope Technologies, Inc.||System and method for visualizing traffic and path in network|
|US7697528 *||Nov 1, 2005||Apr 13, 2010||Nortel Networks Limited||Multilink trunking for encapsulated traffic|
|US7725622 *||Nov 29, 2006||May 25, 2010||Townsend Analytics, Ltd.||Data distribution system and method|
|US7835353 *||Dec 13, 2006||Nov 16, 2010||Tellabs San Jose, Inc.||Apparatus and methods for establishing virtual private networks in a broadband network|
|US7957399||Dec 19, 2008||Jun 7, 2011||Microsoft Corporation||Array-based routing of data packets|
|US8004990||Dec 1, 2009||Aug 23, 2011||Juniper Networks, Inc.||Network traffic distribution across parallel paths|
|US8014411||Dec 21, 2004||Sep 6, 2011||Tellabs San Jose, Inc.||Apparatus and methods for establishing virtual private networks in a broadband network|
|US8031640 *||Oct 2, 2006||Oct 4, 2011||Fujitsu Limited||Packet transmission apparatus, packet forwarding method and packet transmission system|
|US8161185 *||Apr 24, 2006||Apr 17, 2012||Cisco Technology, Inc.||Method and apparatus for assigning IPv6 link state identifiers|
|US8218553 *||Jul 22, 2009||Jul 10, 2012||Juniper Networks, Inc.||Load balancing network traffic on a label switched path using resource reservation protocol with traffic engineering|
|US8259585||Apr 17, 2009||Sep 4, 2012||Juniper Networks, Inc.||Dynamic link load balancing|
|US8300660 *||Jan 30, 2010||Oct 30, 2012||Fujitsu Limited||Transmitting apparatus|
|US8553705||Aug 8, 2011||Oct 8, 2013||Tellabs San Jose, Inc.||Apparatus and methods for establishing virtual private networks in a broadband network|
|US9025615||Sep 10, 2013||May 5, 2015||Tellabs Operations, Inc.||Apparatus and methods for establishing virtual private networks in a broadband network|
|US20040190490 *||Dec 15, 2003||Sep 30, 2004||Alcatel||Device for determining switching paths in a label switched communication network in the presence of selection attributes|
|US20050099947 *||Dec 21, 2004||May 12, 2005||Michael Kazban||Apparatus and methods for establishing virtual private networks in a broadband network|
|US20050232230 *||Apr 14, 2005||Oct 20, 2005||Intec Netcore, Inc.||System and method for visualizing traffic and path in network|
|US20080222273 *||Mar 7, 2007||Sep 11, 2008||Microsoft Corporation||Adaptive rendering of web pages on mobile devices using imaging technology|
|US20100214913 *||Aug 26, 2010||Juniper Networks, Inc.||Load balancing network traffic on a label switched path using resource reservation protocol with traffic engineering|
|US20100238954 *||Jan 30, 2010||Sep 23, 2010||Fujitsu Limited||Transmitting apparatus|
|US20120099456 *||Oct 24, 2011||Apr 26, 2012||Fujitsu Limited||Reception apparatus, communication apparatus, transmission apparatus, receiving method, and transmitting method|
|US20120207012 *||Aug 16, 2012||Juniper Networks, Inc.||Load balancing network traffic on a label switched path using resource reservation protocol with traffic engineering|
|CN101478499B||Jan 8, 2009||Jan 4, 2012||清华大学深圳研究生院||Flow allocation method and apparatus in MPLS network|
|U.S. Classification||370/230, 370/392, 370/400, 370/395.5, 370/395.32, 370/356, 370/235, 370/466, 370/389|
|International Classification||H04L12/66, H04L12/56, H04J3/16, H04J1/16, H04L12/28|
|Cooperative Classification||H04L45/745, H04L45/04, H04L45/50|
|European Classification||H04L45/745, H04L45/04, H04L45/50|
|Nov 12, 2002||AS||Assignment|
Owner name: JUNIPER NETWORKS, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:FERGUSON, DENNIS C.;SHETH, NISCHAL;KUWABARA, KEN;REEL/FRAME:013499/0883;SIGNING DATES FROM 20021007 TO 20021112
|Aug 19, 2008||CC||Certificate of correction|
|Apr 4, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Apr 2, 2015||FPAY||Fee payment|
Year of fee payment: 8